Coral CDN support for URLs with non-standard ports

submitted by Lenny Domnitser

THANKS

@@ -13,6 +13,7 @@ Morten Frederiksen - Support WordPress LinkManager OPML
 Harry Fuecks    - default item date to feed date
 Antonio Cavedoni - Django templates
 Morten Frederiksen - expungeCache
+Lenny Domnitser - Coral CDN support for URLs with non-standard ports
 
 This codebase represents a radical refactoring of Planet 2.0, which lists
 the following contributors:

filters/coral_cdn_filter.py

@@ -3,14 +3,15 @@ Remap all images to take advantage of the Coral Content Distribution
 Network <http://www.coralcdn.org/>.
 """
 
-import sys, urlparse, xml.dom.minidom
+import re, sys, urlparse, xml.dom.minidom
 
 entry = xml.dom.minidom.parse(sys.stdin).documentElement
 
 for node in entry.getElementsByTagName('img'):
     if node.hasAttribute('src'):
         component = list(urlparse.urlparse(node.getAttribute('src')))
-        if component[0]=='http' and component[1].find(':')<0:
+        if component[0] == 'http':
+            component[1] = re.sub(r':(\d+)$', r'.\1', component[1])
             component[1] += '.nyud.net:8080'
             node.setAttribute('src', urlparse.urlunparse(component))
 

tests/data/filter/coral_cdn.xml

@@ -1,7 +1,10 @@
 <entry xmlns="http://www.w3.org/2005/Atom">
   <content>
      <div xmlns="http://www.w3.org/1999/xhtml">
-       <img src="http://example.com/foo.png"/>
+       Plain old image: <img src="http://example.com/foo.png"/>
+       Host has a non-standard port: <img src="http://example.com:1234/foo.png"/>
+       A non-port colon: <img src="http://u:p@example.com/foo.png"/>
+       Several colons: <img src="http://u:p@example.com:1234/foo.png"/>
      </div>
   </content>
 </entry>

tests/test_filters.py

@@ -11,8 +11,11 @@ class FilterTests(unittest.TestCase):
 
         output = shell.run(filter, open(testfile).read(), mode="filter")
         dom = xml.dom.minidom.parseString(output)
-        imgsrc = dom.getElementsByTagName('img')[0].getAttribute('src')
-        self.assertEqual('http://example.com.nyud.net:8080/foo.png', imgsrc)
+        imgsrcs = [img.getAttribute('src') for img in dom.getElementsByTagName('img')]
+        self.assertEqual('http://example.com.nyud.net:8080/foo.png', imgsrcs[0])
+        self.assertEqual('http://example.com.1234.nyud.net:8080/foo.png', imgsrcs[1])
+        self.assertEqual('http://u:p@example.com.nyud.net:8080/foo.png', imgsrcs[2])
+        self.assertEqual('http://u:p@example.com.1234.nyud.net:8080/foo.png', imgsrcs[3])
 
     def test_excerpt_images1(self):
         config.load('tests/data/filter/excerpt-images.ini')